Search This Blog

Powered by Blogger.

Blog Archive

Labels

Footer About

Footer About

Labels

Showing posts with label browser data. Show all posts

Chrome, Edge Extensions Found Stealing Cryptocurrency and Browser Data

 



Nineteen browser extensions for Google Chrome and Microsoft Edge were used to deliver a modular malware framework capable of draining cryptocurrency wallets, stealing credentials and account data, extracting browser history, and displaying ClickFix-style lures.

Application security company Socket identified 18 malicious Chrome extensions and one Microsoft Edge extension as part of a campaign it says may have been operating since February 2024. The extensions contained separate JavaScript modules for different forms of data theft and could receive additional payloads from attacker-controlled infrastructure.

Several of the extensions initially provided the functionality described on their store listings and did not contain malicious code when first published. Socket found that five legitimate extensions were later acquired from their original developers and modified to include the malware. Malicious functionality was then distributed through updates, allowing existing users to receive the altered code without installing a new extension.

One affected extension, “Enable Right Click & Copy — Smart Unlock + OCR,” had at least 70,000 Chrome users when the malicious version appeared. Its Microsoft Edge version had around 10,000 users at the time.

The extensions connect to command-and-control servers through encrypted WebSocket connections. The servers can provide JavaScript modules to the extensions, allowing the operators to add or change functionality without replacing the entire extension. Socket found that the downloaded modules were encrypted before being stored locally and that the framework could use changing C2 infrastructure.

The malware also interferes with protections built into websites. It removes Content Security Policy headers from pages visited by the user, weakening restrictions that can prevent unauthorized scripts and other resources from executing. It then injects malicious JavaScript into websites using hidden HTML elements, giving the attackers a way to interact with pages opened by the victim.

Socket identified 16 malicious modules with different capabilities.

One group of modules targets cryptocurrency wallets across Ethereum Virtual Machine-compatible networks, Solana and Tron. The malware can interfere with legitimate “Connect Wallet” and “Swap” buttons, redirecting transactions or wallet connections to attacker-controlled destinations.

Another module targets users visiting Ledger and Trezor websites. Instead of allowing the legitimate pages to operate normally, the malware can replace them with convincing phishing pages designed to obtain wallet recovery seed phrases. Access to a seed phrase can give an attacker control over the associated cryptocurrency wallet.

The framework also targets cryptocurrency services including Coinbase, Binance, Kraken, OKX, MEXC, KuCoin, Bybit and MetaMask. Its modules can collect session information, authentication tokens, account details and wallet balances from these services.

Credential theft extends beyond cryptocurrency platforms. A separate component can record information entered into forms across websites, including usernames, passwords and other form data. Additional modules collect information associated with Facebook and LinkedIn accounts, while another extracts browser history.

The campaign also incorporates ClickFix-style social engineering. The malware can display fake browser-update messages that instruct users to perform actions on their computers. In the observed implementation, the attacker can place a command in the clipboard and persuade the victim to paste and execute it. This technique shifts part of the attack from automated browser compromise to user-assisted execution.

The extension acquisition strategy adds another layer to the campaign. Browser extensions can receive updates after installation, meaning users who originally downloaded a legitimate utility may later receive altered code. Chrome's extension update process can automatically check for and install updates, so the user does not necessarily need to revisit the Chrome Web Store for the malicious version to reach an existing installation.

Google removed the identified malicious extensions from the Chrome Web Store. Socket also reported the Microsoft Edge extension to Microsoft. At the time of Socket's investigation, however, the Edge version remained available and its operators had changed the C2 domain used by the malware.

Socket said the 16 modules it identified may not represent the framework's complete capabilities. Its modular design allows operators to introduce additional functionality and deploy new payloads through the C2 infrastructure.

Users who installed any of the affected extensions should remove them and treat credentials entered while the extensions were active as potentially exposed. Passwords for affected accounts should be changed, with additional authentication protections enabled where available.

Cryptocurrency users whose wallets may have been exposed should consider the wallet compromised and transfer remaining assets to a newly created wallet with a newly generated recovery phrase.

New Malware “Storm” Steals Browser Data and Hijacks Sessions Without Passwords

 



A newly identified infostealer called Storm has emerged on underground cybercrime forums in early 2026, signalling a change in how attackers steal and use credentials. Priced at under $1,000 per month, the malware collects browser-stored data such as login credentials, session cookies, and cryptocurrency wallet information, then covertly transfers the data to attacker-controlled servers where it is decrypted outside the victim’s system.

This change becomes clearer when compared to earlier techniques. Traditionally, infostealers decrypted browser credentials directly on infected machines by loading SQLite libraries and accessing local credential databases. Because of this, endpoint security tools learned to treat such database access as one of the strongest indicators of malicious activity.

The approach began to break down after Google Chrome introduced App-Bound Encryption in version 127 in July 2024. This mechanism tied encryption keys to the browser environment itself, making local decryption exponentially more difficult. Initial bypass attempts relied on injecting into browser processes or exploiting debugging protocols, but these techniques still generated detectable traces.

Storm avoids this entirely by skipping local decryption. Instead, it extracts encrypted browser files and quietly sends them to attacker infrastructure, removing the behavioural signals that endpoint tools typically rely on. It extends this model by supporting both Chromium-based browsers and Gecko-based browsers such as Firefox, Waterfox, and Pale Moon, whereas tools like StealC V2 still handle Firefox data locally.

The data collected includes saved passwords, session cookies, autofill entries, Google account tokens, payment card details, and browsing history. This combination gives attackers everything required to rebuild authenticated sessions remotely. In practice, a single compromised employee browser can provide direct access to SaaS platforms, internal systems, and cloud environments without triggering any password-based alerts.

Storm also automates session hijacking. Once decrypted, credentials and cookies appear in the attacker’s control panel. By supplying a valid Google refresh token along with a geographically matched SOCKS5 proxy, the platform can silently recreate the victim’s active session.

This technique aligns with earlier research by Varonis Threat Labs. Its Cookie-Bite study showed that stolen Azure Entra ID session cookies can bypass multi-factor authentication, granting persistent access to Microsoft 365. Similarly, its SessionShark analysis demonstrated how phishing kits intercept session tokens in real time to defeat MFA protections. Storm packages these methods into a commercial subscription service.

Beyond credentials, the malware collects files from user directories, extracts session data from applications like Telegram, Signal, and Discord, and targets cryptocurrency wallets through browser extensions and desktop applications. It also gathers system information and captures screenshots across multiple monitors. Most operations run in memory, reducing the likelihood of detection.

Its infrastructure design adds resilience. Operators connect their own virtual private servers to Storm’s central system, routing stolen data through infrastructure they control. This setup limits the impact of takedowns, as enforcement actions are more likely to affect individual operator nodes rather than the core service.

Storm supports multi-user operations, allowing teams to divide responsibilities such as log access, malware build generation, and session restoration. It also automatically categorises stolen credentials by service, with visible rules for platforms including Google, Facebook, Twitter/X, and cPanel, helping attackers prioritise targets.

At the time of analysis, the control panel displayed 1,715 log entries linked to locations including India, the United States, Brazil, Indonesia, Ecuador, and Vietnam. While it is unclear whether all entries represent real victims or test data, variations in IP addresses, internet service providers, and data volumes suggest ongoing campaigns.

The logs include credentials associated with platforms such as Google, Facebook, Twitter/X, Coinbase, Binance, Blockchain.com, and Crypto.com. Such information often feeds into underground credential marketplaces, enabling account takeovers, fraud, and more targeted intrusions.

Storm is offered through a tiered pricing model: $300 for a seven-day trial, $900 per month for standard access, and $1,800 per month for a team licence supporting up to 100 operators and 200 builds. Use of an additional crypter is required. Notably, once deployed, malware builds continue operating even after a subscription expires, allowing ongoing data collection.

Security researchers view Storm as part of a broader evolution in credential theft. By shifting decryption to remote servers, attackers avoid detection mechanisms designed to identify on-device activity. At the same time, session cookie theft is increasingly replacing password theft as the primary objective.

The data collected by such tools often marks the beginning of further attacks, including logins from unusual locations, lateral movement within networks, and unauthorised access patterns.


Indicators of compromise include:

Alias: StormStealer

Forum ID: 221756

Registration date: December 12, 2025

Current version: v0.0.2.0 (Gunnar)

Build details: Developed in C++ (MSVC/msbuild), approximately 460 KB in size, targeting Windows systems


This advent of Storm underlines how cybercriminal tools are becoming more advanced, automated, and difficult to detect, requiring organisations to strengthen monitoring of sessions, user behaviour, and access patterns rather than relying solely on traditional credential protection methods.